Researchers have used a machine-learning model to analyze blood proteins to estimate the biological, as opposed to the chronological, age of bodily organs. It’s a way of predicting an apparently healthy person’s risk of developing conditions like heart failure, Alzheimer’s disease, and diabetes that may lead to earlier treatment.
Chronological age doesn’t always correspond with biological age. The aging process can cause a deterioration in the structure and function of organs that increases the risk of most chronic diseases. While there are many methods by which biological aging can be assessed, most provide a single whole-body measure and not information about the age of particular organs.
Stanford Medicine researchers have led a study that has identified protein biomarkers that can estimate the biological age of an organ and, based on that, an individual’s risk of disease.
“We can estimate the biological age of an organ in an apparently healthy person,” said Tony Wyss-Coray, corresponding author of the study. “That, in turn, predicts a person’s risk for disease related to that organ.”
The researchers looked at 11 organs, organ systems or tissues: heart, lung, brain, kidney, liver, pancreas, intestine, muscle, fat, immune system, and blood vessels (vasculature). They measured 4,979 proteins in 5,676 adults aged 20 to 90 and flagged proteins whose genes were expressed at least four times higher in one organ than any other organ. Of all the proteins measured, 856 (17.9%) met this definition.
A machine-learning algorithm was trained to estimate the chronological age of the 11 organs using the highly expressed proteins as inputs. They also trained an ‘organismal’ model using organ-nonspecific plasma proteins and a ‘conventional’ model using all proteins regardless of specificity to compare the contribution of specific organs to a shared gaining signature. For each individual, the algorithm produced an ‘age gap’, a measure of that individual’s biological age relative to same-aged peers. Previous studies have found an association between age gaps and mortality risk.
The researchers observed that individuals with the same conventional age gap had diverse organ aging profiles, with some individuals having extreme aging in one or more organs relative to the general population.
“When we compared each of these organs’ biological age for each individual with its counterparts among a large group of people without obvious severe diseases, we found that 18.4% of those age 50 or older had at least one organ aging significantly more rapidly than the average,” Wyss-Coray said. “And we found that these individuals are at heightened risk for disease in that particular organ in the next 15 years.”
Having a single accelerated aging organ carried a 20% to 50% higher mortality risk. While only 1.7% of individuals showed extreme aging in multiple organs, their mortality risk was found to be significantly increased.
“They had 6.5 times the mortality risk of somebody without any pronouncedly aged organ,” said Wyss-Coray.
In terms of specific organs, those with accelerated heart aging had a 250% increased risk of heart failure, as well as an increased risk of heart attack and atrial fibrillation. A single standard deviation increase in the brain age gap conferred a 34% increased risk of a clinically relevant increase in cognitive decline over five years. Accelerated brain and vascular aging predicted Alzheimer’s disease progression independently from and as strongly as tau protein levels, the current blood-based biomarker for the disease. There were also strong associations between an extreme-aging kidney – that is, two standard deviations above the norm – and both hypertension and diabetes.
Identifying organs that are aging quickly in apparently healthy people may mean treating them earlier. And the identification of organ-specific proteins that indicate excessive aging and associated disease risk could lead to new targeted drugs.
“The resulting organ aging models can predict mortality, organ-specific functional decline, disease risk and progression and aging heterogeneity between tissues,” said the researchers. “This approach is minimally invasive, requiring only a small blood sample, and could be easily applied to understand the effects of health interventions, such as lifestyle modifications and drug therapies, at the organ level.”
The study was published in the journal Nature.
Source: Stanford Medicine